from __future__ import unicode_literals
+import binascii
import collections
+import email
import getpass
+import io
import optparse
import os
import re
import subprocess
import sys
import itertools
+import xml.etree.ElementTree
try:
except ImportError: # Python 2
import urlparse as compat_urlparse
+try:
+ import urllib.response as compat_urllib_response
+except ImportError: # Python 2
+ import urllib as compat_urllib_response
+
try:
import http.cookiejar as compat_cookiejar
except ImportError: # Python 2
except ImportError: # Python 2
from urllib import urlretrieve as compat_urlretrieve
+try:
+ from html.parser import HTMLParser as compat_HTMLParser
+except ImportError: # Python 2
+ from HTMLParser import HTMLParser as compat_HTMLParser
+
try:
from subprocess import DEVNULL
string = string.replace('+', ' ')
return compat_urllib_parse_unquote(string, encoding, errors)
+try:
+ from urllib.parse import urlencode as compat_urllib_parse_urlencode
+except ImportError: # Python 2
+ # Python 2 will choke in urlencode on mixture of byte and unicode strings.
+ # Possible solutions are to either port it from python 3 with all
+ # the friends or manually ensure input query contains only byte strings.
+ # We will stick with latter thus recursively encoding the whole query.
+ def compat_urllib_parse_urlencode(query, doseq=0, encoding='utf-8'):
+ def encode_elem(e):
+ if isinstance(e, dict):
+ e = encode_dict(e)
+ elif isinstance(e, (list, tuple,)):
+ e = encode_list(e)
+ elif isinstance(e, compat_str):
+ e = e.encode(encoding)
+ return e
+
+ def encode_dict(d):
+ return dict((encode_elem(k), encode_elem(v)) for k, v in d.items())
+
+ def encode_list(l):
+ return [encode_elem(e) for e in l]
+
+ return compat_urllib_parse.urlencode(encode_elem(query), doseq=doseq)
+
+try:
+ from urllib.request import DataHandler as compat_urllib_request_DataHandler
+except ImportError: # Python < 3.4
+ # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py
+ class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler):
+ def data_open(self, req):
+ # data URLs as specified in RFC 2397.
+ #
+ # ignores POSTed data
+ #
+ # syntax:
+ # dataurl := "data:" [ mediatype ] [ ";base64" ] "," data
+ # mediatype := [ type "/" subtype ] *( ";" parameter )
+ # data := *urlchar
+ # parameter := attribute "=" value
+ url = req.get_full_url()
+
+ scheme, data = url.split(':', 1)
+ mediatype, data = data.split(',', 1)
+
+ # even base64 encoded data URLs might be quoted so unquote in any case:
+ data = compat_urllib_parse_unquote_to_bytes(data)
+ if mediatype.endswith(';base64'):
+ data = binascii.a2b_base64(data)
+ mediatype = mediatype[:-7]
+
+ if not mediatype:
+ mediatype = 'text/plain;charset=US-ASCII'
+
+ headers = email.message_from_string(
+ 'Content-type: %s\nContent-length: %d\n' % (mediatype, len(data)))
+
+ return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
+
try:
compat_basestring = basestring # Python 2
except NameError:
except ImportError: # Python 2.6
from xml.parsers.expat import ExpatError as compat_xml_parse_error
+if sys.version_info[0] >= 3:
+ compat_etree_fromstring = xml.etree.ElementTree.fromstring
+else:
+ # python 2.x tries to encode unicode strings with ascii (see the
+ # XMLParser._fixtext method)
+ etree = xml.etree.ElementTree
+
+ try:
+ _etree_iter = etree.Element.iter
+ except AttributeError: # Python <=2.6
+ def _etree_iter(root):
+ for el in root.findall('*'):
+ yield el
+ for sub in _etree_iter(el):
+ yield sub
+
+ # on 2.6 XML doesn't have a parser argument, function copied from CPython
+ # 2.7 source
+ def _XML(text, parser=None):
+ if not parser:
+ parser = etree.XMLParser(target=etree.TreeBuilder())
+ parser.feed(text)
+ return parser.close()
+
+ def _element_factory(*args, **kwargs):
+ el = etree.Element(*args, **kwargs)
+ for k, v in el.items():
+ if isinstance(v, bytes):
+ el.set(k, v.decode('utf-8'))
+ return el
+
+ def compat_etree_fromstring(text):
+ doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+ for el in _etree_iter(doc):
+ if el.text is not None and isinstance(el.text, bytes):
+ el.text = el.text.decode('utf-8')
+ return doc
+
+if sys.version_info < (2, 7):
+ # Here comes the crazy part: In 2.6, if the xpath is a unicode,
+ # .//node does not match if a node is a direct child of . !
+ def compat_xpath(xpath):
+ if isinstance(xpath, compat_str):
+ xpath = xpath.encode('ascii')
+ return xpath
+else:
+ compat_xpath = lambda xpath: xpath
try:
from urllib.parse import parse_qs as compat_parse_qs
nv = name_value.split('=', 1)
if len(nv) != 2:
if strict_parsing:
- raise ValueError("bad query field: %r" % (name_value,))
+ raise ValueError('bad query field: %r' % (name_value,))
# Handle case of a control-name with no equal sign
if keep_blank_values:
nv.append('')
return ord(c)
+compat_os_name = os._name if os.name == 'java' else os.name
+
+
if sys.version_info >= (3, 0):
compat_getenv = os.getenv
compat_expanduser = os.path.expanduser
# The following are os.path.expanduser implementations from cpython 2.7.8 stdlib
# for different platforms with correct environment variables decoding.
- if os.name == 'posix':
+ if compat_os_name == 'posix':
def compat_expanduser(path):
"""Expand ~ and ~user constructions. If user or $HOME is unknown,
do nothing."""
userhome = pwent.pw_dir
userhome = userhome.rstrip('/')
return (userhome + path[i:]) or '/'
- elif os.name == 'nt' or os.name == 'ce':
+ elif compat_os_name == 'nt' or compat_os_name == 'ce':
def compat_expanduser(path):
"""Expand ~ and ~user constructs.
else:
compat_getpass = getpass.getpass
-# Old 2.6 and 2.7 releases require kwargs to be bytes
+# Python < 2.6.5 require kwargs to be bytes
try:
def _testfunc(x):
pass
if err is not None:
raise err
else:
- raise socket.error("getaddrinfo returns an empty list")
+ raise socket.error('getaddrinfo returns an empty list')
else:
compat_socket_create_connection = socket.create_connection
from tokenize import generate_tokens as compat_tokenize_tokenize
__all__ = [
+ 'compat_HTMLParser',
'compat_HTTPError',
'compat_basestring',
'compat_chr',
'compat_cookiejar',
'compat_cookies',
+ 'compat_etree_fromstring',
'compat_expanduser',
'compat_get_terminal_size',
'compat_getenv',
'compat_itertools_count',
'compat_kwargs',
'compat_ord',
+ 'compat_os_name',
'compat_parse_qs',
'compat_print',
'compat_shlex_split',
'compat_urllib_parse_unquote',
'compat_urllib_parse_unquote_plus',
'compat_urllib_parse_unquote_to_bytes',
+ 'compat_urllib_parse_urlencode',
'compat_urllib_parse_urlparse',
'compat_urllib_request',
+ 'compat_urllib_request_DataHandler',
+ 'compat_urllib_response',
'compat_urlparse',
'compat_urlretrieve',
'compat_xml_parse_error',
+ 'compat_xpath',
'shlex_quote',
'subprocess_check_output',
'workaround_optparse_bug9161',